2012 | OriginalPaper | Buchkapitel
Identification Number Recognition Based on Random Forests
verfasst von : Hua Shi, Cuihua Li, Yi Xie, Quan Zou
Erschienen in: Software Engineering and Knowledge Engineering: Theory and Practice
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
The recognition of identification (ID) numbers in ID card is very important in real applications. In this paper, we obtain the region of interest by using the minimum cross-entropy in spatial domain and determine the type of ID card according to the position of face. ID numbers are separated into different characters by vertical projection, and the character patches are recognized based on Random Forests. The experimental results have shown that the proposed algorithm with low complexity in computation has a high accuracy rate in recognizing the first-generation ID card, the second-generation ID card and the driving ID card. Moreover, this algorithm is robust to the changes of illumination, rotation and image distortion.